16:01:26 #startmeeting Horizon 16:01:26 Meeting started Tue Mar 4 16:01:26 2014 UTC and is due to finish in 60 minutes. The chair is david-lyle. Information about MeetBot at http://wiki.debian.org/MeetBot. 16:01:27 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 16:01:30 The meeting name has been set to 'horizon' 16:01:40 Hello everyone 16:01:42 hello 16:01:46 o/ 16:01:52 hello \o 16:01:53 hello 16:01:55 hello o/ 16:02:10 Hi 16:02:19 hi all 16:02:23 greetings! 16:02:29 hi 16:02:43 The final day for features and strings in Icehouse is upon us 16:02:49 https://launchpad.net/horizon/+milestone/icehouse-3 16:02:57 1 Blocked, 17 Needs Code Review, 22 Implemented 16:03:12 The top two in the list are FFE material 16:03:16 hi 16:03:32 2-3 were approved and had merge conflicts 16:03:46 and some need reviews or more work 16:04:20 if we can clearly identify the items we think need more work and should be pushed to Juno, it would simplify the rest of the day :) 16:04:38 david-lyle, te 16:05:03 david-lyle, the separation of horizon and dashboard is one thing for juno 16:05:04 IMO at this stage, if the "more work" is a nice to have, or smaller bugs that can be handled before RC (e.g. small performance improvements, docs, etc) it's ok to +2 and open a new bug for the extra work 16:05:51 david-lyle: ive had some reviews get approved but out of order (ie, a string of dependent patches), so there are merge conflicts that wont resolve until the dependency chain is reviewed 16:07:25 jomara, I think the ones not approved have -1's on them 16:07:42 do they?? i will double check, from my dashboard they appeared green butr maybe that is misleading 16:07:45 we can discuss it those are enough to block the change and be fixed in the RC 16:08:00 a +2 can hide -1s 16:08:10 david-lyle: oh boy, it looks like if someone +2s, the -1 doesnt show up on from the front view 16:08:17 yeah.. dang 16:08:36 jomara: https://review.openstack.org/#/c/73433/ is just a merge conflict as well 16:08:41 :( IIRC some of them could easily be done after FF and before the RC, e.g. fixing up docstrings 16:08:44 no dependency 16:09:03 I would be fine with that 16:09:13 jomara: seems like the bottleneck is here: https://review.openstack.org/#/c/71395/4 16:09:18 Lot's of good work to block on a doc string 16:09:32 Got approved but it was not merged, I'll approve it again to trigger the build. 16:09:56 lcheng_: ah, thanks 16:10:01 ok, ill try to get everything fixed up ASAP 16:10:12 sorry about that, i didnt realize i wasnt seeing the -1s 16:10:33 lcheng_: The review it depends on isn't merged though, I don't think it'll go through yet 16:10:47 I think jpich is correct 16:11:13 jpich: you're right 16:11:18 https://review.openstack.org/#/c/66603/19 16:11:21 david-lyle: the one you linked is also dependent, but it says its a merge conflict, i assumed it was due to earlier patches 16:12:01 ok, just not a dependency according to gerrit 16:12:22 jomara: I think this is the top of the chain https://review.openstack.org/#/c/66603/19 16:12:59 lcheng_: yes, that should be 16:13:06 lcheng_: it has 2 -1s that were hidden by the +2 16:13:14 so ill start there and owrk my way down 16:14:34 as a heads up https://blueprints.launchpad.net/horizon/+spec/django-1point6 is blocked by an openstack/requirements approval 16:15:16 once that happens, I going to update and release django_openstack_auth and the update Horizon to have a tox env for 1.5 and 1.6 16:15:20 david-lyle: Will adding a django15 job be done separately or is there a related patch already? 16:15:29 Sounds good 16:15:58 in openstack_auth, it will happen in one patch, in Horizon I will update the requirements and add the tox job in the same patch 16:16:14 jomara: thanks, me and david-lyle are in the US. We can follow-up on the heat patches for the rest of the day. 16:16:16 depending on who's awake, I may just push the changes through 16:16:30 lcheng_: awesome, thanks 16:16:34 It's really just book keeping at that point 16:19:35 also looks like https://review.openstack.org/#/c/61032/ is a priority for Hyper-V support 16:20:06 it wasn't targeted until yesterday 16:21:43 So after icehouse-3 is tagged, master will still be icehouse until RC1 is but in a few weeks 16:21:58 then the items left over can start merging to Juno 16:23:36 #topic Open Discussion 16:23:48 I think we were there, but now it's official :) 16:24:12 I worked with sayalilunkad last week to put together a design for Sparklines on the instance table. We are hoping to hear from lsmola on his thoughts on direction for her to go in…here is the design: http://people.redhat.com/~lsurette/OpenStack/Sparklines%20on%20Instance%20Table 16:25:03 you know, that table could really use more columns ;) 16:25:07 haha 16:25:12 lblanchard: great 16:25:22 might push us towards a "Add/Remove Columns" feature :) 16:25:35 lblanchard: nice! 16:25:40 lblanchard: sayali is investigating whether the meters are good for this 16:25:49 but the thoughts is that RAM and vCPU would be two very nice metrics to show here for each instance 16:25:55 lsmola: great! 16:26:16 Maybe we should move the actions toward the beginning of the table at some point, to avoid having to scroll to do stuff 16:26:21 I do like the look of it, very clean 16:26:26 and also…adding two sparklines to the instance table would be a nice attainable goal before her internship ends on the 10th :) 16:26:40 10th of? 16:26:47 david-lyle: march 16:26:50 :( 16:26:55 david-lyle: yeah :( 16:26:58 jpich: great point 16:27:27 or perhaps a different way to render colums 16:27:30 columns 16:27:35 david-lyle: yes for sure 16:27:50 one thing to ask too is…are all of these columns necessary here 16:27:57 or would some be okay to drop to the details view 16:28:02 put state information in the same column but in multiple rows, sparklines in the same column, etc 16:28:39 david-lyle: right, we can definitely merge and reformat some of these values 16:28:55 I will brainstorm an idea for next week 16:29:21 but for now and as dar as sayali is concerned, the layout you have proposed looks like a great way to move forward and have a tangable endpoint 16:29:32 s/dar/far/ 16:29:40 lblanchard: great 16:29:41 great, thanks! 16:30:10 lsmola: will you follow up with her when she comes online? I know she was looking to hear from you on this to confirm. 16:30:31 lblanchard: we have talked today 16:30:33 I'd love to get some attention on the host aggregates patch if some of the cores have a chance to take a look. I reviewed it a few times, not sure if it's ok to leave a final +2 since I made the last change (a couple of tests) -> https://review.openstack.org/#/c/71061/ . It'd be neat to have in Icehouse :) 16:30:38 lsmola: perfect 16:31:32 lblanchard: she just need to make sure the meters gives us correct data and this can be done 16:31:55 lsmola: sounds great, I'll hope they fit in :) 16:32:11 jpich, just rebasing, I would say it's fine to +2 16:32:54 david-lyle: I added a couple of tests as well 16:32:55 :O 16:33:09 well, that's just bonus points :) 16:33:16 hi all. I've a question regarding a blueprint I'm currently working on (https://blueprints.launchpad.net/horizon/+spec/download-images-and-snapshots). project dashboard implementation is ready and I'm currently addressing finishing adding support on the admin dashboard as requested in review.o.o which should be ready today. only UT missing as requested by mrunge. any chance of approving this for I-3 or is it too late now? 16:33:26 heh :) 16:35:30 cgoncalves: It really help to set the milestone target when working on a blueprint so it can be considered during release tracking 16:35:59 lblanchard david-lyle -- in looking at the instance table, if we can drop some of the columns (and put them in details) that would be great, e.g. why do we need to include image name, size, key pair, ip address 16:36:11 otherwise especially when approaching the end of a milestone it's unlikely to get as much review attention 16:36:13 jpich: I did set it, but david-lyle removed it very recently (today morning I think) 16:36:33 16:37:43 cgoncalves, I did, I went through the bps targeted for icehouse and items that had not received much attention or looked like they had work left (adding tests) were moved out 16:37:43 side Q sorry.. how do you set the milestone? I've never been able to edit that field.. bugs or bp 16:38:05 julim: yes, I'm wondering something similar…what is important to see at this table level and what could be put into a details view. Perhaps we could do some user testing here to make sure we make informed changes. 16:38:20 absubram: The assignee should be able to set the target on a blueprint, IIUC 16:38:31 sounds good lblanchard. 16:39:10 jpich: thanks.. I'll keep an eye on it next time I open a bp :) 16:40:27 david-lyle: I didn't look yet for how to implement a UT for testing the correctness of downloading an image from horizon in client-side. in case it's feasible and somehow trivial, I should implement it in time 16:41:31 david-lyle: the request for such UT was only asked a couple hours ago. I've been addressing all comments on review.o.o as fast as I can as you can see (https://review.openstack.org/#/c/74799/) 16:41:41 absubram: Ok! It's part of the steps for creating a blueprint so please say something if that's not actually possible ( https://wiki.openstack.org/wiki/Blueprints#Creation ) 16:42:18 cgoncalves, I'll look at it again a client side download test may not make sense in unit tests 16:42:58 david-lyle: thanks. that was what I understood from mrunge's comment, a client side download test. 16:43:17 jpich: thanks for the link! 16:43:47 absubram: I am a link factory, at your service :) 16:43:49 jpich, they show up targeted all the time, so I hope someone is setting it :) 16:44:48 Maybe someone wrote a prediction bot to save people's time 16:44:52 david-lyle: haha yes.. I've only opened on bp myself (in grizzly!).. it's normally bugs that I open and that's more of what I was thinking of 16:45:00 one* 16:45:46 jpich: haha.. that'd be a nice bot to have! 16:47:54 ah no, I was just for something more checking than just watching if the menu item was added cgoncalves 16:48:21 david-lyle: about my bp - https://blueprints.launchpad.net/horizon/+spec/neutron-subnet-mode-support… neutron side is still sitting on -1 review :(.. am tracking it daily.. I think amotoki is also a reviewer there 16:48:55 will send you a note to let you know if they somehow get their code in 16:49:08 absubram, thanks for the update, I have that bp slated as a possible FFE, may be on the neutron side too, not sure 16:49:26 keep me posted 16:49:33 will do 16:50:10 I had a comment in my review about the edit subnet portion not working (wanted to talk to amotoki about that since he also has a comment in that piece of code) 16:50:28 but its looking like neutron might not support edit in i.. only in J 16:50:51 so even if we got our Horizon support in.. it would only be to add the two new ipv6 attributes during subnet create 16:50:59 mrunge: ok, thanks for your input here! do you happen to have a concrete idea of how that should be tested? any pointers? :-) 16:51:07 we would have to fix update subnet in J 16:51:57 cgoncalves, look at the other tests, like launching machines, creating volumes... 16:52:36 mrunge: ok, will do. 16:52:55 cgoncalves, and if there's anything, feel free to drop me a note 16:53:39 mrunge: sure. thanks once again! 16:54:28 on a separate note though.. I have a bunch of minor bug fix patches out in review and looking for reviewers :p.. I know everyone's trying to get the BP's in.. so whenever people get a chance.. (amotoki hehe) 16:55:06 bugs will take a much higher priority after today 16:55:10 bug fixes can be merged in till the RC, I think people will focus on these more once feature freeze has passed 16:55:11 yeah 16:55:20 stabilisation FTW 16:55:27 oh yes! 16:55:28 yep.. figured so.. it's fine :) thanks 16:55:56 I can only encourage folks to install different environments to test! 16:57:50 if we have time for open Qs.. I did have a question about one particular unit test that's giving me a little trouble on a separate issue.. the launch_instance_post 16:57:58 if not I can send out an email to the mailer 16:58:51 absubram, just post it in the regular room 16:58:58 I don't think we have time here 16:59:02 cool.. will do 16:59:05 that's fine :) 16:59:58 I'd like to thank everyone for the review efforts, core and non-core. We've merged a lot of great work in icehouse and have a lot more ready for Juno. If your bp doesn't make it into icehouse, I will target it for Juno as soon as it opens up. I think the release cycle is just here to add a huge amount of stress every 6 months. And, I suppose to keep us honest :) Thanks everyone. 17:00:26 Have a great week! 17:00:30 #endmeeting